Locomotion in extinct giant kangaroos: were sthenurines hop-less monsters?

Abstract

Sthenurine kangaroos (Marsupialia, Diprotodontia, Macropodoidea) were an extinct subfamily within the family Macropodidae (kangaroos and rat-kangaroos). These "short-faced browsers" first appeared in the middle Miocene, and radiated in the Plio-Pleistocene into a diversity of mostly large-bodied forms, more robust than extant forms in their build. The largest (Procoptodon goliah) had an estimated body mass of 240 kg, almost three times the size of the largest living kangaroos, and there is speculation whether a kangaroo of this size would be biomechanically capable of hopping locomotion. Previously described aspects of sthenurine anatomy (specialized forelimbs, rigid lumbar spine) would limit their ability to perform the characteristic kangaroo pentapedal walking (using the tail as a fifth limb), an essential gait at slower speeds as slow hopping is energetically unfeasible. Analysis of limb bone measurements of sthenurines in comparison with extant macropodoids shows a number of anatomical differences, especially in the large species. The scaling of long bone robusticity indicates that sthenurines are following the "normal" allometric trend for macropodoids, while the large extant kangaroos are relatively gracile. Other morphological differences are indicative of adaptations for a novel type of locomotor behavior in sthenurines: they lacked many specialized features for rapid hopping, and they also had anatomy indicative of supporting their body with an upright trunk (e.g., dorsally tipped ischiae), and of supporting their weight on one leg at a time (e.g., larger hips and knees, stabilized ankle joint). We propose that sthenurines adopted a bipedal striding gait (a gait occasionally observed in extant tree-kangaroos): in the smaller and earlier forms, this gait may have been employed as an alternative to pentapedal locomotion at slower speeds, while in the larger Pleistocene forms this gait may have enabled them to evolve to body sizes where hopping was no longer a feasible form of more rapid locomotion.

Figure 1. Reconstruction of Sthenurus stirlingi.

By Brian Regal.

Figure 2. Skeletons of (A) Sthenurus stirlingi and (B) Macropus giganteus.

Modified from Wells and Tedford, 1995. Original artist Lorraine Meeker, American Museum of Natural History (reproduced here by permission).

Figure 3. Pelvis.

(A) Simosthenurus occidentalis (SAM: P17358). (B) Dendrolagus dorianus (SAM: M9190) (C). Macropus robustus (SAM: M3695). Left lateral view. Scale bar  = 5 cm.

Figure 4. Femur.

(A) Simosthenurus occidentalis (SAM: P17259). (B) Macropus sp. (SAM: P17270). All left side: upper  =  proximal articular view; lower  =  lateral distal view. Scale bar  = 5 cm.

Figure 5. Tibia.

(A) Procoptodon goliah (NMV2010). (B) Macropus giganteus (AMNH 2390). All left side: upper  =  distal articular view (plantar side upwards); lower  =  posterior (plantar) view, showing articulation with tarsus. Scale bar  = 5 cm.

Figure 6. Astragalus and calcaneum.

(A) Simosthenurus occidentalis (SAM: P17258 [reversed]). (B) Macropus giganteus (SAM: P17523 [reversed]). All left side: upper  =  proximodorsal view; lower  =  lateral view. Scale bar  = 2 cm.

Figure 7. Fourth metatarsal.

(A) Sthenurus stirlingi (AMNH 117496). (B) Macropus giganteus (AMNH 2390 [reversed]). All left side: upper  =  proximal articular view; lower  =  distal articular view (plantar side downwards). Scale bar  = 2 cm.

Figure 8. Measurements used in analyses-1.

Drawings, all of left side elements, primarily from photographs of Macropus fuliginosus, AMNH 2390. (A) Pelvis, lateral view. (B) Pelvis, ventral view. (C) Pelvis, dorsal view. (D) Femur, anterior view. (E) Femur, posterior view. (F) Femur, proximal articular view (anterior of shaft downwards). (G) Distal femur, medial view. (H) Distal femur, lateral view. A detailed description of the measurements is provided in Table S2.

Figure 9. Measurements used in analyses-2.

Drawings, all of left side elements, primarily from photographs of Macropus fuliginosus, AMNH 2390, Calcaneum from Macropus giganteus, AMNH 74753). (A) Tibia, lateral view (fibula removed). (B) Tibia, proximal articular view (plantar side downwards). (C) Tibia, distal articular view (plantar side upwards). (D) Astragalus, anterior (dorsal) view. (E) Astragalus, medial view. (F) Astragalus, lateral view. (G) Calcaneum, anterior (dorsal) view. (H) Calcaneum, posterior (plantar) view. (I) Calcaneum, lateral view. (J) Calcaneum (head only) medial view. (K) Calcaneum, distal articular view (plantar side upwards). (L) Pes, anterior (dorsal) view. A detailed description of the measurements is provided in Table S2.

Figure 10. Scaling of long bone length versus diameter (i.e., robusticity).

(A) Femur length versus average femur cross-sectional diameter (B) Tibia length versus average tibia midshaft cross sectional diameter. The regression line and its 95% confidence interval (dotted lines) are also shown (A r² = 0.96, B r² = 0.933). Key: Open diamond  =  Hypsiprymnodon moschatus; filled circles  =  potoroines; open circles  =  extant species of Macropus; half tone circles  =  extinct (“giant”) species of Macropus (M. titan or M. ferragus); filled diamonds  =  extant macropodines (other than Macropus or Dendrolagus) and lagostrophines; open squares  =  species of Dendrolagus; filled stars  =  sthenurines; open stars  =  other extinct taxa. The regression line and its 95% confidence interval (dotted lines) are also shown.

Figure 11. Multivariate analyses of hindlimb bones.

Key as for Figure 10: taxa included and explanation of any composite fossils are listed and explained in Tables S3, 4. (A) Principal Components Analysis using all hind limb bones. The dotted line indicates the division between extant taxa that are more specialized hoppers, and those that are less specialized or that rarely hop. (B) Principal Components Analysis without the pelvis. The dotted line indicates the division between extant taxa that are more specialized hoppers, and those that are less specialized or that rarely hop. The placement of taxa along both components is very similar to that shown in Figure 11A, except where otherwise noted. (C). Discriminant Analysis using all hind limb bones. Sthenurines and extant macropodines (plus Lagostrophus) only. (D). Principal Components Analysis, calcaneum only. Key as for Figure 10 except for the following additions: half tone diamonds  =  extinct Miocene macropodine Dorcopsoides; half tone stars  =  Miocene sthenurines ( =  Hadronomas puckridgi unless otherwise indicated). (E). Discriminant Analysis, calcaneum only. Key as for (D).